Fiachra MAC GÓRÁIN

MICROCOSM AND THE VIRGILIAN PERSONA

I wish to approach the persona of from the perspective of microcosm, which I broadly define as a literary figure involving a comparison between something great and something small. I shall focus on three related microcosmic passages, one from each of the three works: a speech of Tityrus in the first Eclogue, a simile from the fourth Georgic, and the shield of . Thus my discussion will offer something of relevance to the poet’s persona in and across the three works1. I begin with an overview of microcosm in Virgil before exploring the connection between microcosm and the poet’s persona. Virgil’s oeuvre is microcosmic in different ways. The late-antique Neoplatonist commentator Macrobius explicitly compared Virgil to the cosmos2. This comparison is a central reference point for scholarly appreciations of the universal and all-embracing qualities of Virgil’s poetry, especially the Georgics and the Aeneid3. Critics since the fourth-century commentators Aelius Donatus and Servius have noted both the intertextual imperialism of Virgil’s synthetic poetics and the way in which the three works taken together seem to encompass the entire trajectory of human civilization, from pastoralism through agriculturalism to urban polity. The three works also contain many instances of mise en abyme, which is microcosm of a second order. Lucien Dällenbach, author of the classic study on the subject defines mise en abyme as “any internal mirror that reflects the whole of a narrative by simple, repeated or ‘specious’ (or paradoxical) duplication”4. But the term is often applied more loosely, and in Virgil’s case it is useful to distinguish between intertextual and intratextual miniaturizations. The most obvious example of intertextual mise en abyme is the way in which the incorporates a reworking of both Homeric poems. On a smaller scale Book Three of the Aeneid mirrors the embedded narrative of the wanderings of Odysseus5. It has also been argued that the sixth Eclogue is a miniaturization of Callimachus’ Aetia, that the murals on Dido’s temple to present a synopsis of the entire archaic epic tradition, and that the parade of heroes and the shield of Aeneas recapitulate in different ways the Annales of Ennius6. Among the intratextual mises en abyme are the many summaries, proleptic or analeptic, partial or precise, of the works in which they are embedded7. Other passages and whole

1 Scholarly precedents for unitary readings of Virgil’s tripartite corpus include the following: F. Klingner, ‘Die Einheit des vergilischen Lebenswerkes’, Römische Geisteswelt, Munich, 1961, p. 274-9; C. Hardie, ‘The Georgics: A Transitional Poem’, Abingdon, 1971; E. Theodorakopoulos, ‘Closure: the Book of Virgil’, The Cambridge Companion to Virgil, ed. C. Martindale, Cambridge, p. 155-165; P. Hardie, Virgil, Oxford, [Greece and Rome New Surveys in the Classics No. 28], 1998, p. 1, and D. Nelis, ‘From didactic to epic: Georgics 2.458-3.48’, Latin Epic and Didactic Poetry, ed. M. Gale, Swansea, 2004, p. 73-107. 2 On the basis of both style and content; cf. Sat. 1.16.12 omnium disciplinarum peritus; Comm. Somn. Scip. 1.6.44 nullius disciplinae expers, and 1.15.12 disciplinarum omnium peritissimus. Cf. also Sat. 1.24.10-21, Sat. 5.1.7 and 5.1.18-20. 3 Cf. V. Pöschl, ‘Virgil als universaler Dichter’, Lebendige Vergangenheit. Abhandlungen und Aufsätze zur Römischen Literatur. Kleine Schriften III. Herausgegeben von Wolf-Lüder Liebermann, Heidelberg, 1995, p. 73-89, and P. R. Hardie, Virgil’s Aeneid: Cosmos and Imperium, Oxford, 1986, p. 22-5, 33-50, and Ch. 7. 4 L.Dällenbach, The Mirror in the Text, Cambridge, 1989, p. 36, originally published as Le récit spectaculaire, Seuil, 1977. 5 Virgil metapoetically signals this repetition towards the end of the book. A Barchiesi, Journal of Roman Studies 86, 1996, p. 231 points to talia monstrabat relegens errata retrorsus | litora Achaemenides (3.690-1), where relegens errata might on a first reading be interpreted as ‘rereading the Wanderings.’ Cf. renarrabat (3.717). Cf. also D. Quint, Epic and Empire, Princeton, 1993. 6 Cf. J. J. Clauss, ‘Vergil's Sixth Eclogue: The Aetia in Rome’, Hellenistica Groningana VI: Callimachus, ed. M. A. Harder et al., Groningen, p. 71-93 on the sixth Eclogue, within which of course Silenus offers a digest of cosmological epic (31-40); for Dido’s murals cf. A. Barchiesi, ‘Rappresentazioni del dolore e interpretazione nell'Eneide’, Antike und Abendland 40, p. 109-24; for the parade of heroes cf. P. Hardie, The Epic Successors of Virgil, Cambridge, 1993; for the shield of Aeneas cf. P. Hardie, Cosmos, and A. Barchiesi, ‘Virgilian narrative: Ecphrasis’, The Cambridge Companion to Virgil, ed. C. Martindale, Cambridge, 1997, p. 271-81. 7 Some examples: lines 70-1 from the sphragis of the tenth Eclogue offer a characterization of the collection, haec sat erit, diuae, uestrum cecinisse poetam, | dum sedet et gracili fiscellam texit hibisco, as Servius recognized in his note on 71, allegoricos autem

1 books have been read as mirroring either the contours of the surrounding work, or some of its more important thematic thrusts8. Even the epic’s incipit, Arma uirumque is a microcosm both intra- and intertextual. Insofar as Arma uirumque is the poem’s title, Arma uirumque cano means both ‘I sing of arms and the man’ and also ‘I am singing the Aeneid.’ But the phrase also points to the poem’s status as a microcosm of both Homeric epics9. Microcosm offers a privileged locus for accessing the poet’s persona. First of all many of these microcosmic passages occur within the poet’s direct self-representations and are deeply implicated in his literary and generic self-positioning. Secondly, many others are focalized through an internal narrator or surrogate author, and in all of these cases there is either a clear echo of words in the poet’s voice or of some part of his work, or a miniaturizing allusion to one or more of his models. Surrogate authors are often considered refractions of the poet’s persona, just as internal audiences or viewers may be considered surrogate external readers. Fowler, Hardie and others have discussed analogies between acts of composition and reading within the poem, and our reaction to the poem as external readers. Fowler in particular has addressed the special connection between mise en abyme and authorial surrogacy10. The present opusculum situates itself within this scholarly tradition by considering three related microcosmic passages which are fundamental to an understanding of Virgil’s literary, generic and ideological negotiations. I would argue that Virgil’s microcosmic preoccupations can be traced back to the first Eclogue, to the second speech of Tityrus, who may be read as a surrogate author. The intermittent identification of Tityrus with the authorial voice goes back to Servius (Comm. Buc. 1.1 hoc loco Tityri sub persona Vergilium debemus accipere, non tamen ubique sed tantum ubi exigit ratio), and it has been a pervasive idea, though not universally accepted, in the history of criticism11. But of course the Eclogues are famous for their obfuscation of the authorial voice, and Tityrus is only one of many herdsmen who could be considered authorial surrogates. He responds as follows to Meliboeus’ enquiry about the identity of the god who has allowed him to keep his lands (19-25):

significat se composuisse hunc libellum tenuissimo stilo. The first four lines of the Georgics programme the content of the four books; cf. Servius ad loc. Cf. R. A. B. Mynors, Virgil. Georgics, Oxford, 1990 and S. J. Harrison, ‘Laudes Italiae (Georgics 2.136-175): Virgil as a Caesarian Hesiod’, Patria diversis gentibus una? Unita politica e identita etniche nell’Italia antica, ed. G. Urso, Milan, 2008, p. 231-42, on Geo. 2.143-8 and on the Laudes Italiae (Geo. 2.136-76) and the sphragis at Geo. 4.559-60. Cf. the Sibyl’s prophecy of the war in Latium at Aen. 6.83-94, which finds echoes in the poet’s voice at 7.41-5. L. Bocciolini Palagi, La trottola di Dioniso, Bologna, 2007, p. 191 reads ’ words at Aen. 10.41, Allecto medias Italum bacchata per urbes, as a kind of a summary of Allecto’s three furious visitations in book 7. 8 Some examples: cf. D. Hershkowitz, ‘The Aeneid in Aeneid 3’, Vergilius, 37, 1991, p. 69-76 and D. Quint, Epic and Empire, on Aeneid 3; cf. G. K. Galinsky, ‘Aeneid V and the Aeneid’, American Journal of Philology, 89, 1968, p. 157-85; cf. D. Fowler, ‘Epic in the Middle of the Wood: Mise en Abyme in the Nisus and Euryalus Episode’, Intratextuality: Greek and Roman Textual Relations, ed. A. Sharrock and H. Morales, Oxford, 2000, p. 89-113; cf. P. Hardie, Cosmos, p. 53-66 and D. Nelis, Vergil's Aeneid and the Argonautica of Apollonius Rhodius, Leeds, 2001, p. 99-112 on the song of Iopas; cf. P. Hardie, Cosmos, p. 326 on Aen. 6.740-743; cf. G. K. Galinsky, ‘The Hercules-Cacus Episode in Aeneid VIII’, AJP, 87, 1966, p. 18- 51 and Ll. Morgan, ‘Assimilation and Civil War: Hercules and Cacus: Aeneid 8’, Vergil’s Aeneid: Augustan Epic and Political Context, ed. H. P. Stahl, Swansea, 1998, p. 175-98; cf. P. Hardie, Cosmos, p. 83-4 and Ll. Morgan, Patterns of Redemption in Virgil’s ‘Georgics’, Cambridge, 1999, p. 94-6 on the song of Clymene (Geo. 4.345-7). 9 On the titular force of a poem’s incipit cf. Servius, Comm. Aen. 1. Proem., veteres incipiebant carmen a titulo carminis sui; Ecl. 5.85-7, arma uirumque in Ov. Tr. 2.534 and Mart. Ep. 8.55.19 and 14.185; cf. G. B. Conte, The Rhetoric of Imitation, Ithaca and London, 1986, p. 70-86, and W. Levitan, ‘Give Up the Beginning? Juno’s Mindful Wrath (Aeneid 1.37)’, Liverpool Classical Monthly, 18, 1993, p. 14. 10 Surrogate authors: cf. D. Fowler, ‘Epic in the Middle’, p. 29-30, and P. Hardie, Virgil, 75-9; on Aeneas ‘misreading’ the murals on Dido’s temple cf. C. G. Perkell, Reading Vergil's Aeneid: An Interpretive Guide, Norman, OK, 1999, p. 45-6; on Iopas, who quotes from Georgics Book Two, cf. P. Hardie, Cosmos, p. 52-66; on Aeneas and the poet cf. Pöschl, ‘Universaler Dichter’, p. 80 and A. M. Bowie, ‘Aeneas Narrator’, Proceedings of the Virgil Society, 26, p. 41-51; on cf. A. Barchiesi, ‘Ecphrasis’ and S. Casali, ‘The Making of the Shield: Inspiration and Repression in the Aeneid’, Greece and Rome, 53, 2006, p. 185-204; on Cretheus in Aen. 9. 774-7 cf. P. Hardie, Virgil. Aeneid IX, Cambridge, 1994, p. 238-9. 11 Tityrus as Virgil: cf. M. C. J. Putnam and J. M. Ziolkowski, The Virgilian Tradition, New Haven and London, 2008, index, s. v. ‘Tityrus, Virgil as’. Commentators who have read Tityrus as a figure for the poet include La Cerda (1612), Heyne- Wagner (1830), Conington (1872), and Page (1937). For challenges to the identification cf. I. M. Le M. Du Quesnay, ‘Vergil’s First Eclogue’, Papers of the Liverpool Latin Seminar, 3, p. 32-8 and B. W. Breed, Pastoral Inscriptions, London, 2006, p. 102-3.

2 Vrbem quam dicunt Romam, Meliboee, putaui stultus ego huic nostrae similem, quo saepe solemus pastores ouium teneros depellere fetus. sic canibus catulos similes, sic matribus haedos noram, sic paruis componere magna solebam. uerum haec tantum alias inter caput extulit urbes quantum lenta solent inter uiburna cupressi.

The city they call Rome, Meliboeus, I thought – stupid me – was like ours here, where we shepherds so often drive the tender young of our flocks. Thus I knew puppies were like dogs, and kids like their dams, thus I was in the habit of comparing great things with small. But this one has reared her head as high among all other cities as cypresses often do among the bending guelder-rose bushes. (tr. Fairclough-Goold, adapted)

Tityrus comes across as charmingly naïve, and the circuitous form and rustic analogies of his response typify the simplicity and nonchalance of the bucolic world at its happiest12. He attempts a full-blown simile, but for Pöschl and Rieks, his failure to get beyond the kind of correlative comparison (tantum ... quantum) which is the norm in the Eclogues comes to express the incomparability of Rome13. Perhaps more importantly, with its first words urbem quam dicunt Romam, and its comparisons between great and small, Tityrus’ speech springs the poem’s frame of reference from the green cabinet to the city which impinges on it. This builds on the hints of a wider world already present several lines earlier in patriae ... patriam (3-4), and looks forward to the extension of the poem’s spatial embrace (both horizontal and vertical) to include earth, sea and sky, and the boundaries of the known world14. Tityrus’ speech thus indicates one of the poem’s central themes, the relationship between the small and hitherto peaceful settlement of the idealized countryside and the expanding reach of the imperial metropolis. This opposition between Rome and the idyll is central to Pöschl’s universalizing interpretation of the poem15. In short, Tityrus’ words figure the first Eclogue as a microcosm. I suggest that Tityrus’ speech lends itself to a metapoetic reading. As he enlarges on Rome, he effectively offers three similes, or fledgling similes (urbem quam dicunt Romam, Meliboee, putaui | [...] huic nostrae similem [...] sic canibus catulos similes, sic matribus haedos | noram), which since his visit to the city he now sees – stultus ego – are inadequate to the task. These are followed by a comment on the practice of composing similes (sic paruis componere magna solebam), and finally by the more ambitious simile which comes closer to the mark (uerum haec tantum alias inter caput extulit urbes | quantum lenta solent inter uiburna cupressi). The confrontation between Rome and the pastoral idyll, between great and small, necessitates for Tityrus a change in the way pastoral poetry is composed. As Servius observed, the first three similes involve magnitudinis differentia, whereas the fourth involves et generis et magnitudinis differentia. Rome is of a different order altogether, nam est sedes deorum, as Tityrus can now appreciate16. All of this has obvious relevance for Virgil writing about recent history and contemporary matters. My claim of a metapoetic reading derives some support from the fact that the Latin word simile may carry its technical sense at least as early as Cicero17. Equally the range of meanings of the verb componere includes

12 Servius on 1.19; cf. V. Pöschl, Die Hirtendichtung Virgils, Heidelberg, 1964, p. 35 and E. W. Leach, Vergil's Eclogues: Landscapes of Experience, Ithaca, NY, 1974, p. 124. 13 Pöschl, Hirtendichtung, p. 36; R. Rieks, ‘Die Gleichnisse Vergils’, Aufstieg und Niedergang der römischen Welt, ii 31.2, p. 1051. 14 Cf. 59-69, Tityrus: Ante leues ergo pascentur in aethere cerui | et freta destituent nudos in litore piscis, | ante pererratis amborum finibus exsul | aut Ararim Parthus bibet aut Germania Tigrim, | quam nostro illius labatur pectore uultus. | Meliboeus: At nos hinc alii sitientis ibimus Afros, | pars Scythiam et rapidum cretae ueniemus Oaxen | et penitus toto diuisos orbe Britannos. | en umquam patrios longo post tempore finis | pauperis et tuguri congestum caespite culmen, | post aliquot, mea regna, uidens mirabor aristas? 15 Pöschl, ibidem. 16 Servius on 1.22. 17 OLD s.v. simile 2 cites Rhet. Her. 2.46, Cic. Fin. 3.46, Or. 3.163, Tusc. 2.13 and Quint. Inst. 5.11.34. Cf. H. Lausberg, Handbook of Literary Rhetoric, Leiden, p. 200-2 and 377-80, where simile is often synonymous with similitudo in ancient

3 ‘compose’ as well as ‘compare’18. Paruis componere magna is a particularly loaded phrase in a collection which thematizes the antithetical relationship between ‘grand’ and ‘humble’ genres of poetry, and the subject matter which they may treat19. This tension (as I see it) is most apparent in the change of key at the beginning of the fourth Eclogue, heralded by the poetological invocation (Sicelides Musae, paulo maiora canamus!) and the following lines which plead for the worthiness of pastoral poetry to negotiate consular themes (non omnis arbusta iuuant humilesque myricae; | si canimus siluas, siluae sint consule dignae). Such an elevation of pastoral to consular status could also be glossed as paruis componere magna.A similar generic tension surfaces in the recusatio which begins the sixth Eclogue, where the distinction seems to be between martial epic and pastoral. The recusatio is intratextually conversant20 with the beginning of the fourth Eclogue (6.1-5):

Prima Syracosio dignata est ludere uersu [~ 4.1 Sicelides; 4.3 dignae] nostra neque erubuit siluas habitare, Thalea. [~ 4.3 siluas siluae; 4.1 Musae] cum canerem reges et proelia, Cynthius aurem [~ 4.1 canamus; 4.1 maiora] uellit, et admonuit: ‘Pastorem, Tityre, pinguis [~ 4.1 maiora] pascere oportet ouis, deductum dicere carmen.’

My Muse first deigned to sport in Sicilian strains, and blushed not to dwell in the woods. When I was fain to sing of kings and battles, the Cynthian plucked my ear and warned me: “A shepherd, Tityrus, should feed sheep that are fat, but sing a lay fine-spun.” (tr. Fairclough-Goold)

Virgil is dramatizing a mandatory regression from martial epic (reges et proelia) back to pastoral, and the terms in which the distinction is expressed (pinguis ... ouis, deductum ... carmen) also look back to (the same?) Tityrus’ comparison of great things with small in the first Eclogue21. To use Servius’ terminology, between martial epic and pastoral, there exists et magnitudinis et generis differentia. Thus once again the Eclogues are microcosmic in the way that they retrospectively afford a vista onto the epic which Virgil went on to write22. Turning now from pastoral to the grander genre of didactic epic, my argument for a metapoetic reading of Tityrus’ speech gains some retrospective support from Virgil’s use of the phrase si parua licet componere magnis, an inversion of Tityrus’ phrase, in the simile in the fourth book of the Georgics which compares the bees to Cyclopes (4.170-7):

ac ueluti lentis Cyclopes fulmina massis cum properant, alii taurinis follibus auras accipiunt redduntque, alii stridentia tingunt aera lacu; gemit impositis incudibus Aetna; illi inter sese magna ui bracchia tollunt in numerum uersantque tenaci forcipe ferrum: non aliter, si parua licet componere magnis, Cecropias innatus apes amor urget habendi

rhetorical texts. Quintilian classifies Tityrus’ canibus catulos and matribus haedos as similes (Inst. 5.11.30). 18 OLD s.v. compono 8a cites instances of the verb used to mean composing poetry: Cic. Mur. 26; Ov. Tr. 5.12.60. For a metapoetic reading of componere in an apertural context at Aen. 1.274 cf. D. Nelis, ‘From didactic to epic’, p. 95. 19 For Latin and Greek parallels to paruis componere magna cf. A. Otto, Die Sprichwörter und sprichwörtlichen Redensarten der Römer, Leipzig, 1890, § 1008. 20 D. Nelis, ‘From didactic to epic’, p. 74, with further references, notes these and other parallels between the fourth and sixth Eclogues. 21 On Virgil’s allusion to Callimachus’ Aetia prologue cf. W. Clausen, ‘Callimachus and Latin Poetry’, Greek, Roman and Byzantine Studies, 5, 1964, p. 181-96; J. Farrell, Vergil's Georgics and the Traditions of Ancient Epic, Oxford, 1991, p. 295-300; R. F. Thomas, ‘Callimachus Back in Rome’, Callimachus. Hellenistica Groningana Vol. 1, ed. M. A. Harder et al., Groningen, 1993, p. 197-215; and A. Cameron, Callimachus and His Critics, Princeton, 1995, p. 454-75. 22 As we have noted there is also allusion to cosmological epic at the beginning of the song of Silenus; cf. Ecl. 6.31-40.

4 munere quamque suo.

And as, when the Cyclopes in haste forge bolts from tough ore, some with oxhide bellows make the blasts come and go, others dip the hissing brass in the lake, while Aetna groans under the anvils laid upon her; they, with mighty force, now one, now another, raise their arms in measured cadence, and turn the iron with gripping tongs – even so, if we may compare small things with great, an inborn love of gain spurs on the Attic bees, each after its own office. (tr. Fairclough-Goold)

Here the phrase is in the poet’s voice, and it has explicit poetological force23. It is also deeply embedded in a nexus of micro- and macrocosms, which I believe works on three levels. The first of these is the way in which the hive is a paradigm for the Roman state24. The bees’ microcosmic status is evident from the opening lines of the book, and there has already been extensive play on the contrast between large and small. As with Tityrus’ comparison, this play works both on a stylistic and an ideological level25. Virgil draws attention to the microcosm with the phrase si parua licet componere magnis, and by comparing the bees’ labour with ‘the grandest industrial spectacle available to the imagination of Antiquity, the busy forging of ’s thunderbolts’26. The second way in which the bees-Cyclopes simile is microcosmic is that it resonates intertextually with the micro- and macrocosmic strategies of Plato’s Republic. The idea that Virgil’s account of apian society owes something to the discussion of the ideal state in Plato’s Republic is one that has been around at least since Servius; I believe it has much to commend it27. Plato’s Socrates is interested in justice in the soul, and he initiates discussion of the ideal state in order to discern justice ‘writ large’28. As such, the ideal state in the Republic is a macrocosm. Thus a shared feature of Plato’s macrocosm and Virgil’s microcosm is that they are both concerned with government and with social order. But within Plato’s macrocosm there is an extended hive metaphor, which, as Adam notes, ‘is worked out with unusual completeness, even for Plato’29. It implicates the hive, the king-bee, and (most prominently) drones30. It is ingenious that Plato’s macrocosm contains an embedded apian microcosm. I believe that Virgil alludes to this embedded microcosm, first of all by aligning the social organization of his bees with that in Plato’s ideal state, and secondly by including a Cyclopean macrocosm within his own apian microcosm. The inversion of Tityrus’ sic paruis componere magna solebam in the phrase si parua

23 The poet of the Georgics styles himself a shepherd at 3.286-7, and invites a unitary reading of both Eclogues and Georgics in the sphragis of the later poem. 24 Synoptic discussions of this complex issue include J. Griffin, ‘The Fourth Georgic, Virgil, and Rome’, G&R, 26, 1979, p. 61-80; D. Nelis, ‘The Aristaeus episode and Aeneid 1’, From Erudition to Inspiration. A Booklet for Michael. Essays in Honour of M. J. McGann, Belfast, p. 3-18; and Ll. Morgan, Patterns of Redemption. The comparison of bees with humans was of course not a Virgilian innovation; cf. Il. 2.87, 12.167; Hes. Th. 594ff., WD 302-6; Varro, RR, 3.16.6; Cic. Off. 1.157. 25 Cf. Geo. 4.3-6, 19, 20, 26, 38, 55, 68, 76, 79, 83, and R. F. Thomas, Virgil. Georgics III-IV, Cambridge, 1982, on 4.6. 26 R. A. B. Mynors, Georgics, p. 280. 27 Servius on Geo. 4.153; cf. J. L. de la Cerda, P. Virgilii Maronis Bucolica et Georgica, Lyons, 1612, p. 470, on Geo. 4.153, and L. P. Wilkinson, The Georgics of Virgil, Cambridge, 1969, p. 176. Thomas, Georgics, on 4.153-5 ascribes the common features to the ethnographical tradition. I see the following points of contact: communality of offspring and dwelling places at Geo. 4.153-4 and Pl. Resp. 5.457c-d; educunt at Geo. 4.163 and education at Resp. 2. 376b-412b; guardianship, cf. fu/lakej at Resp. 2.374d8 and Geo. 4.165 custodia; social organization by lot, cf. Geo. 4.165 sorti with (e.g.) Resp. 5.460a8- 10 and 8.557a. 28 Resp. 2.368d-369a. 29 J. Adam, The Republic of Plato, Cambridge, 1902, on 8.554b. 30 On the apian metaphor in the Republic cf. D. Tarrant, ‘Imagery in Plato's Republic’, Classical Quarterly, 40, 1946, p. 33-4; A. Pelletier, ‘L'image du «frelon» dans la République de Platon’, Revue de philologie, de littérature et d'histoire anciennes, 22, 1948, p. 131-46; and R. S. Liebert, ‘Apian Imagery and the Critique of Poetic Sweetness in Plato’s Republic’, Transactions of the American Philological Association, 140, 2010. The Philosopher Kings are encouraged, against their will, to behave like leaders and ‘king-bees in a hive’ at 7.520b; all other instances of the metaphor are drone-focussed: the oligarchical man at 8.552c (cf. also 554b, 554d); the transition to democracy (555d-559d); at 8.564c the doctor or law-giver, like a good bee-keeper must subdue or eliminate the drones; the transition from democracy to tyranny (564d-567d); the psychic condition of the tyrannical man (9.573a).

5 licet componere magnis points to the double inversion of Plato in the simile in Book Four of the Georgics. The significance of the Platonic intertextuality is more difficult to determine. Because Plato’s apian metaphor is most sustained with reference to drones, particularly in the discussion of degenerate forms of government in Books Eight and Nine, one might almost say that it is a drone-metaphor with several references to bees (industrious bees, that is) rather than the other way around. This emphasis is conspicuous in light of the favourable comparisons between bees and humans elsewhere in the Platonic corpus31. On the one hand this might seem to offer grounds for a ‘pessimistic’ reading of Georgics Book Four: it is possible that even after the bugonia has provided a new hive, the persistence of drones may signal a recidivist tendency to social degeneration. On the other hand, Virgil directly addresses the threat posed by drones in the Georgics (4.244 immunisque sedens aliena ad pabula fucus), and it is significant that the bees have mechanisms to suppress them (4.167-8 aut agmine facto | ignavum fucos pecus a praesepibus arcent). The third microcosmic level in the bees-Cyclopes simile is also intertextual. Here Virgil looks back, by means of window-allusion, to ’s Odyssey through Callimachus’ Hymn to Artemis. His intertextual juxtaposition of a grand epic with a smaller one could reasonably be described as parua componere magnis (or vice versa). Joseph Farrell has admirably exposed the allusive subtlety of the passage32. Its closest model is evidently the scene in Callimachus’ hymn where Artemis finds the Cyclopes toiling away at a me¢ga eãrgon (Hymn 3.46-61):

auåqi de£ Ku¢klwpaj meteki¢aqe: tou£j me£n eãtetme nh¢s% eni£ Lipa¢r$ (Lipa¢rh ne¢on, a¦lla£ to¢t' eãsken ouãnoma¢ oi¥ Meligouni¢j) e¦p' aãkmosin ¥Hfai¢stoio e¥stao¢taj peri£ mu¢dron: e¦pei¢geto ga£r me¢ga eãrgon: i¥ppei¢hn tetu¢konto Poseida¢wni poti¢strhn. 50 ai¥ nu¢mfai d' eãddeisan, oàpwj iãdon ai¦na£ pe¢lwra prho¢sin ¦Ossai¢oisin e¦oiko¢ta, pa¤si d' u¥p' o¦fru¢n fa¢ea mouno¢glhna sa¢kei iãsa tetraboei¢ deino£n u¥poglau¢ssonta) kai£ o¥ppo¢te dou¤pon aãkousan aãkmonoj h¦xh¢santoj e¦pi£ me¢ga poulu¢ t' aãhma 55 fusa¢wn au¦tw¤n te baru£n sto¢non: auåe ga£r Aiãtnh, auåe de£ Trinakri¢h Sikanw¤n eàdoj, auåe de£ gei¢twn ¦Itali¢h, mega¢lhn de£ boh£n e¦pi£ Ku¢rnoj a¦u+¢tei, euåq' oiàge r¥aisth¤raj a¦eira¢menoi u¥pe£r wãmwn hä xalko£n zei¢onta kamino¢qen h¦e£ si¢dhron 60 a¦mboladi£j tetu¢pontej e¦pi£ me¢ga muxqi¢sseian.

And straightaway she went to visit the Cyclopes. Them she found on the isle of Lipara – Lipara in later times, but back then its name was Meligunis – at the anvils of Hephaestus, standing round a molten mass of iron. For a great work was being hastened on: they fashioned a horse- trough for Poseidon. And the were frightened when they saw the terrible monsters that resembled the crags of Ossa: all had single eyes beneath their brows, like a shield of fourfold hide for size, glaring terribly from under; and when they heard the din of the anvil echoing loudly, and the great blast of the bellows and the heavy groaning of the Cyclopes themselves. For Aetna cried aloud, and Trinacia cried, the seat of the Sicanians, their neighbour Italy cried, and Cyrnos added a mighty cry too, when they lifted their hammers above their shoulder and smote with rhythmic swing the bronze glowing from the furnace of iron, labouring greatly. (tr. A. W. Mair, slightly updated)

31 R. S. Liebert, ibidem, cites Meno 72b Phd. 82b and Pol. 301e. 32 The present paragraph is much indebted to J. Farrell, Vergil’s Georgics, p. 243-5.

6 There is extensive play on great and small in this hymn, especially in the passage quoted. Artemis’ nymphs take fright at the terrifying size and din of the whole scene (note the preponderance of words indicating great size, great din etc., underlined), but Artemis herself, though still little, is not cowed33. The Cyclopes’ me¢ga eãrgon turns out to be nothing grander than a horse-trough for Poseidon. Bornmann and Casali have ascribed metapoetic significance to the deflationary tone of poti¢strhn in line 50, and to the asymmetrical abruptness with which the hoplopoia is finally concluded34. Callimachus alludes in this passage to the blinding of Polyphemus in Book Nine of the Odyssey35. Virgil in turn alludes both to Callimachus and through him to Homer36. The window-allusion thus directs the reader from a small-scale example of Hellenistic epos back to a grander one. This may be read as closely related to the increasing density of Homeric allusion in the Georgics, as the poem proceeds, particularly in Book Four37. I would argue that the Georgics simile builds on Tityrus’ phrase from the Eclogues to look forward to the Aeneid38. Each of the three microcosmic levels in the simile reflects a transition from small to large. His bees pass along a similar trajectory. The first (not especially marked) reference to bees comes in the first Eclogue, in Meliboeus’ idealized vignette of the pastoral idyll which Tityrus may continue to enjoy (54), while Meliboeus himself must migrate to the boundaries of the empire (64-8). In the Georgics, as we have seen, the bees have grander significance. Then there are various ways in which the heroic pursuits of the bees in Georgics Book Four prefigure the Aeneid, not least the fact that four of the main peoples in the later poem are compared to bees, in language that persistently echoes, sometimes very closely, the bees of Georgics Four39. This can only corroborate allegorical readings of the bees in the Georgics. But our bees-Cyclopes simile in particular is repeated almost verbatim in Book Eight of the Aeneid40. This implies a significant relationship between the three passages which I discuss, Tityrus’ speech, the simile, and the shield of Aeneas. Let us now turn to Book Eight of the Aeneid. As Vulcan prepares to fashion armour for Aeneas, he orders the Cyclopes to interrupt their work on Jupiter’s thunderbolt with the following words (8.439-41):

‘tollite cuncta’ inquit ‘coeptosque auferte labores, Aetnaei Cyclopes, et huc aduertite mentem: arma acri facienda uiro.

“Away with all!” he cries. “Remove the tasks you have begun, Cyclopes of Aetna, and turn your thoughts to this! Arms for a brave warrior you must make.... (tr. Fairclough-Goold)

The mandate to craft arms for Aeneas, arma acri facienda uiro, can hardly fail to recall the poem’s incipit, with all its titular force41. As such the shield may be read as a kind of surrogate Aeneid, and Vulcan as a

33 Cf. Hymn 3.72-80, which leads into Artemis’ bold address to the Cyclopes. 34 F. Bornmann, Callimachi Hymnus in Dianam, Florence, 1964, p. 29, 44, and S. Casali, ‘The Making of the Shield’, p. 199- 200. 35 Od. 9.389-93, pa¢nta de¢ oi¥ ble¢far' a¦mfi£ kai£ o¦fru¢aj euâsen a¦u+tmh£ | glh¢nhj kaiome¢nhj: sfarageu¤nto de¢ oi¥ puri£ r¥i¢zai. | w¥j d' oàt' a¦nh£r xalkeu£j pe¢lekun me¢gan h¦e£ ske¢parnon | ei¦n uàdati yuxr%= ba¢pt$ mega¢la i¦a¢xonta | farma¢sswn: 36 Cf. J. Farrell, Vergil’s Georgics, p. 244: ‘Vergil carefully alludes to the element that Callimachus borrows from Homer (alii stridentia tingunt | aera lacu 172-173). By doing so, he discloses the literary history of the motif.’ 37 On this phenomenon cf. J. Farrell, Vergil’s Georgics, Ch. 6, and Ll. Morgan, Patterns of Redemption. 38 On the idea of the Georgics as a transitional poem cf. n. 1, above, especially D. Nelis, ‘From didactic to epic’. 39 Carthaginians at 1.430-436; Romans at 6.707-9; Trojans at 7.64-7; and Latins at 12.587-92. 40 The simile from Geo. 4 is repeated almost verbatim at Aen. 8.449-53, but the whole episode which concludes with these lines (416-453) also reworks the passages quoted from Callimachus’ Hymn to Artemis and Odyssey 9, as well as Il. 18.372- 89, Hes. Theog. 140, Apoll. Arg. 1.730-4, and Lucr. DRN 6.246-378; cf. S. Casali, ‘The Making of the Shield’, p. 197-203. 41 Virgil also uses labor (~ labores, Aen. 8.439) of his own poetic endeavour at Ecl. 10.1, Geo. 2.39, 3.288, 4.6, 116; and the metapoetic resonance of mens (Aen. 8.440) is a lynch-pin of D. Fowler’s argument (‘Epic in the Middle’, p. 99) for Nisus as a surrogate author (Aen. 9.187).

7 surrogate Virgil42. The shield has Augustus’ Actian victory at its centre (675 in medio), and in a later panel Augustus surveying the triumphal procession from the temple of Palatine (720). Thus the shield may be read as a fulfilment of Virgil’s pledge in the proem of Georgics Book Three to compose an epic poem on the exploits of the princeps, with Caesar in the middle (16 in medio mihi Caesar erit templumque tenebit)43. But if the shield is a surrogate Aeneid, it is a miniature one, and in fact there are numerous other ways in which the shield is microcosmic. It is one shield to counter all the Latin enemies’ weapons (8.447-8 ingentem clipeum informant, unum omnia contra | tela Latinorum); in this respect it coheres with the synecdochic schema of one-for-all, which is one of the Aeneid’s fundamental structuring principles, and also a microcosmic pattern44. But metapoetically speaking, it is also unum omnia contra in the sense that it reworks and subsumes an impressive range of intertextual models, which are themselves microcosms. This is true of the Aeneid in general, but of the shield in particular. Si parua licet componere magnis, it is modelled on Homer’s shield of Achilles, which Hardie, following the ancient scholia, has discussed as an imago mundi, and which is also in some respects a mise en abyme of the Iliad45. Another of its models is Jason’s cloak in the Argonautica, which is itself a stylistic and thematic mise en abyme of Apollonius’ epic, and which the ancient scholiasts also read as a cosmological allegory46. As we have noted, it is a recapitulation of the Annales of Ennius: it begins with , proceeds chronologically (629 in ordine), and the part of its historical content which is narrated to us extends to the poet’s own day. The shield’s Ennianism, its annalistic mode, reflects for Barchiesi ‘a kind of antagonistic poetics, a road not taken’47. There is, then, (to return once again to Servius’ analysis of Tityrus’ speech) et magnitudinis et generis differentia between the shield, a miniature Aeneid, and the Aeneid itself. Coterminous with the (updated) temporal expanse of Ennius’ Annales the shield is an elliptical précis of Roman history, but what exactly is depicted on it? Let us examine how it is introduced (624- 9):

tum leuis ocreas electro auroque recocto, hastamque et clipei non enarrabile textum. illic res Italas Romanorumque triumphos haud uatum ignarus uenturique inscius aeui fecerat ignipotens, illic genus omne futurae stirpis ab Ascanio pugnataque in ordine bella.

...then the smooth greaves of electrum and refined gold, the spear, and the shield’s ineffable (?) fabric. There the story of Italy and the triumphs of Rome had the Lord of Fire fashioned, not unversed in prophecy or unknowing of the age to come; there, every generation of the stock to spring from Ascanius, and the wars they fought in their sequence. (tr. Fairclough-Goold)

The phrase non enarrabile textum suggests that there is more on the shield than can be fully narrated, and that the panels which follow this introduction represent selections. This view is supported by Virgil’s own description of the shield: he tells us that Vulcan had fashioned on it genus omne futurae | stirpis ab Ascanio. The words genus omne figure the shield as an icon of infinity, of the imperium sine fine which Jupiter prophesied in Book One (279). Its literary aesthetic seems identical to the Callimacheanism of Cornelius Nepos, as praised by Catullus in his first and programmatic poem (5-7

42 Cf. S. Casali, ‘The Making of the Shield’, p. 200. 43 This is only one possible view; on the fulfilment of Virgil’s pledge in the proem to Geo. 3 cf. D. Nelis, ‘From didactic to epic’, with further references. 44 Cf. P. Hardie, Epic Successors, p. 27-32. 45 P. Hardie, ‘Imago Mundi: Cosmological and Ideological Aspects of the Shield of Achilles’, Journal of Hellenic Studies, 105, 1985, p. 11-31; id., Cosmos, Ch. 8; and O. Taplin, ‘The Shield of Achilles within the Iliad’, G&R, 27, p. 1-21. 46 Apoll. Arg. 1.730-68; cf. D. Nelis, Aeneid and Argonautica, p. 345-59 and S. Goldhill, The Poet’s Voice, Cambridge, 1991, p. 308-11. 47 A. Barchiesi, ‘Ecphrasis’, p. 275; on the shield’s Ennianism cf. also D. Nelis, ‘From didactic to epic’, p. 92.

8 iam tum, cum ausus es unus Italorum | omne aeuum tribus explicare cartis | doctis, Iuppiter, et laboriosis), the difference being that Virgil’s shield encompasses the future as well as the past48. Through Vulcan’s divine artistry, Virgil represents himself as a universal poet, as a poet of infinity. One of Dällenbach’s categories of mise en abyme is the presence within a work of ‘repeated’ or ‘infinite’ duplication. He illustrates this with a quotation from Derrida. ‘When one can read a book within the book, an origin within the origin, a centre within the centre, this leads us into an abyss [abîme], a bottomless and infinite duplication’49. There are several examples of this on the shield of Aeneas, which point once again to its status as an icon of infinity. First, on the shield which has come down from heaven (608-9), Vulcan had devised shields fallen from heaven (664-5 lapsa ancilia caelo | extuderat). Secondly, Alexander McKay has argued that all the locations mentioned on the shield are to be found along the route of the pompa triumphalis50. And the shield depicted, we are told, res Italas Romanorumque triumphos (626). So if the series of tableaux on the shield points to a triumphal parade, then Augustus’ triple triumph (714-28) is a triumph within a triumph. Further layers of recursion emerge when the reader looks at Aeneas looking at a shield which depicts Augustus reviewing gifts, perhaps including shields (721-2 dona recognoscit populorum aptatque superbis | postibus)51. We are in a microcosmic hall of mirrors. Now that we have followed an intratextual line from small to large, and from the beginning to (not quite) the end, it is time to draw some conclusions. Virgil’s self-representations are rarely direct, and even those that appear to be so are intensely stylized and metaliterary. It is impossible to draw any distinctions between what might be autobiographical and what is a literary motif. So while most readers would find it highly unlikely that he had planned out his poetic career at the time of writing the Eclogues, the composition of a long epic poem about reges et proelia was clearly on the horizon of the metafictional persona which we can glimpse behind Tityrus, indeed when he is at his most recusational and Callimachean. It is certainly true that my argument depends heavily on a retrospective reading; but Virgil too was a retrospective reader as well as a retrospective poet, and the intratextual cues between the three passages which I have discussed invite a unitary reading of the tripartite corpus. Under the guise of Tityrus (Tityri sub persona) Virgil uses the microcosmic analogy to thematize the relationship between small and grand genres of poetry, and to explore how humble pastoral may comprehend Rome and the universal. As ever, genre and ideology are in close dialogue with one another. This is no less true of the Georgics, where Virgil harks back to Tityrus, and retains his preoccupations with the relationship between small and large. In the bees-Cyclopes simile he alludes to models which themselves thematize this relationship, some of which have extensive political import. If the microcosmic patterns of Georgics Book Four seemed vertiginous, those in Book Eight of the Aeneid are more complex still. Here Virgil uses microcosmic tropes (especially different levels of mise en abyme) to accord the shield of Aeneas a privileged status within the epic, to indicate his subordination of the entire epic tradition, and to figure himself as a poet of infinity. Virgil’s use of microcosm allows us to perceive his generic ascent from small to large, from pastoral through didactic to the composition of an epic that would rival all its predecessors, and that served as a model for all its successors. His persona is literary and metaliterary, and his concerns are universal52.

48 The augmentor of the Servian commentary envisaged as infinite the temporal span of the poem which Virgil promises in the proem to Georgics Book Three. On Geo. 3.48 Tithoni prima quot abest ab origine Caesar he comments: ab infinito infinitum, quia Tithoni origo non potest comprehendi. For D. Nelis, ‘From didactic to epic’, p. 88, this places Caesar in medio in an additional sense. Auden’s poem, Secondary Epic, offers a relevant and insightful reading of Virgil’s shield. 49 Cf. J. Derrida, De la grammatologie, Paris, Minuit, 1967, p. 437, ‘quand on peut lire un livre dans le livre, une origine dans l’origine, un centre dans le centre, c’est l’abîme, le sans-fond du redoublement infini.’ (Eng. tr. B. Johnson, London, 1981.) 50 A. G. McKay, ‘Non enarrabile textum? The Shield of Aeneas and the Triple Triumph in 29 BC (Aen. 8.630-728)’, Virgil’s Aeneid: Augustan Epic and Political Context, ed. H. P. Stahl, Swansea, 1998, p. 199-222. 51 A. Barchiesi, ‘Ecphrasis’, p. 276: “The ‘gifts’ to be affixed to the doorposts would be, typically, shields”. Cf. also M. C. J. Putnam, Virgil’s Epic Designs: Ekphrasis in the Aeneid, New Haven and London, 1998, p. 119-88. 52 I extend my warmest thanks to the following for their generous and enriching responses to earlier incarnations of the ideas presented here: S. Alexander, F. Budelmann, A. Dale, P. Hardie, D. Kennedy, R. S. Liebert, M. Marshall, and D. O’Rourke.

9 BIBLIOGRAPHY

S. Casali, ‘The Making of the Shield: Inspiration and Repression in the Aeneid’, G&R, 53, 2006, p. 185- 204. L. Dällenbach, The Mirror in the Text, Cambridge, 1989. D. Fowler, ‘Epic in the Middle of the Wood: Mise en Abyme in the Nisus and Euryalus Episode’, Intratextuality: Greek and Roman textual Relations, ed. A. Sharrock and H. Morales, Oxford, 2000, p. 89- 113. P. Hardie, Virgil’s Aeneid: Cosmos and Imperium, Oxford, 1986. R. A. B. Mynors, Vergili Opera, Oxford, 1969. D. Nelis, ‘From didactic to epic: Georgics 2.458-3.48’, Latin Epic and Didactic Poetry, ed. M. Gale, Swansea, 2004, p. 73-107. V. Pöschl, ‘Virgil als universaler Dichter’, Lebendige Vergangenheit. Abhandlungen und Aufsätze zur Römischen Literatur. Kleine Schriften III. Herausgegeben von Wolf-Lüder Liebermann, Heidelberg, 1995, p. 73-89.

10